How To Optimize Your Email Marketing Campaigns Through Testing
from Ron Evans
Marketing gurus have already established that it is not possible to maximize the results of your marketing campaigns. This is because you never know which new and clever trick will bring an ever better result than what has gone before. Email marketing, like any other form of direct marketing, follows this essential rule.However, it is possible to optimize your marketing campaign results, meaning to measurably improve the results you are getting now relative to the results you had in the past. This applies to all forms of marketing, direct marketing, and advertising. The best way to do this is to test different variations on the message, the creative, the timing, and the frequency.
MESSAGE
For example, all marketing messages begin with a headline that communicates an offer or benefit. "Learn more about solar power heating for your home" and "Get a free solar powered radio with any solar heating system" are both possible messages for a company that sells solar equipment, but clearly are quite different in their approach. And its not just the big pitch, but also small subtle variations on wording that can make a big difference. As Mark Twain said, "The difference between the right word and the almost right word is like the difference between lightning, and a lightning bug."
CREATIVE
The look and feel, color choices, design layout, and other aesthetics are the creative elements for an email marketing campaign. Should a specific campaign be bright and edgy, or sober and business-minded? Should you use mostly pictures, or lots of text? Managers responsible for bottom-line business activities tend to respond to text and interactive emails from people they know and trust. On the other hand, managers responsible for more creative activities (i.e. marketing, advertising, design) and consumers respond best to highly creative email campaigns.
TIMING
The time of day or day of week can have a big impact on the results for your email marketing campaigns. So is Tuesday afternoon or Thursday morning better for your particular offer? You can only find out by testing. Please read our article "Timing Is Everything: When and How Often To Use Email Marketing For Best Results" for a more detailed discussion of this topic.
FREQUENCY
How often and how many times that you make contact, can significantly influence the results of your email marketing campaigns. Should you send once a week for three weeks, or every two weeks? Only testing can tell. Typically, rules of direct marketing show that "hitting" your target audience multiple times will ensure a higher response rate. The "Timing Is Everything" article also talks about the topic of frequency for your email campaigns.
Preparing To Test
One of the most powerful marketing testing techniques is the concept of the "split run". The idea is a very simple one that dates from long before the Internet. Basically, you segment your list into multiple groups, and then try different variations on each group. Once you determine which variation draws the better response, you can improve your results by incorporating these "lessons learned" into future campaigns.
Segmenting Your Lists
Marketers who have large enough lists can split off 10-20% of the total addresses to use just for testing. Whatever yields the best results will be used on the remaining 80-90% of the list. Those with smaller lists will have to do testing on the full list, and may not get to test as many variations. The more factors that you want to test, the more complex it becomes to segment your list, while avoiding sending to the same person twice, or to recipients that have opted off of your list at some point during the testing.
For example, lets say we want to test three different factors within a single mailing: creative "A" and "B", message "F" and "G", and timing "X" and "Y". We would need to first divide the list into "A" and "B" groups. Then divide the "A" group into "AF" and "AG" groups, and the "B" group into "BF" and "BG" groups. Lastly, we would need to divide the "AF" group into the "AFX" and "AFY" groups, the "AG" group into the "AGX" and "AGY" groups, the "BF" group into the "BFX" and "BFY" groups, and the "BG" group into the "BGX" and "BGY" groups (see Figure 1).
Figure 1
Whew! These three letter designated groups are the ones that we are going to use for our testing.
Interpreting The Results
Determining which variation performed better is simply a matter of comparing all of the sibling test group results with each other. In the example above, how did "AFX", "AFY", "AGX", "AGY", "BFX", "BFY", "BGX", and "BGY" compare to each other in number of opens? How about click-thrus? Make sure that you have a large enough sampling of recipients that have been exposed to each variation or you may not be able to draw solid conclusions about how effective your campaign has been. Remember, you can test factors for a single mailing as well as testing factors over multiple mailings.
Conclusion
Only once you have started to measure the effectiveness of your email marketing campaigns, can you begin to optimize the results. Split testing can help you get information much more quickly. When you test and analyze the results you can determine what is working well, or not, and use that information to improve your bottom line.
